Skip to content

Python script for object detection using Intel Movidius Neural Compute Stick and a pretrained model to be further used on Raspberry-based hexapod robot.

Notifications You must be signed in to change notification settings

kapiblue/ncs-object-detection

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

15 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

ncs-object-detection

Python script for object detection using Intel Movidius Neural Compute Stick and a pretrained model.

Installation

I tested the code on VirtualBox with Raspberry Pi Desktop. I installed Version 1 of the NCS SDK according to this tutorial https://www.bouvet.no/bouvet-deler/adding-ai-to-edge-devices-with-the-movidius-neural-compute-stick. This is also an useful guide for getting started.

Getting started

The code is based on the security-cam app from ncapzoo https://github.com/movidius/ncappzoo/tree/ncsdk1

Run main.py for person detection of the sample image included in the /images folder. Feel free to modify the script, for example change the model (remember to update also the labels file, dimensions and the color mode).

python main.py

Technologies

  • Python 2.7
  • Intel NCS SDK
  • MobileNet SSD for object detection
  • VirtualBox with Raspberry Pi Desktop

Sample image: freepik.com

About

Python script for object detection using Intel Movidius Neural Compute Stick and a pretrained model to be further used on Raspberry-based hexapod robot.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages